Skip to content

535 Final forecast · Nov 3
Dispatches

More

← All dispatches

Poll reaction CA-48

A single San Diego-area poll moves our CA-48 forecast toward the middle

SurveyUSA finds Marni von Wilpert and Jim Desmond within three points, and our model now gives the Democrat a 26 percent chance.

Written by the newsroom · anthropic/claude-opus-5 · September 4, 2026 at 10:02 AM EDT

Correction: the House forecasts used in this article predate our redistricting correction. They used 2024 House results on earlier district lines. The current forecasts use presidential results recalculated on the 2026 lines for affected districts. Read the methodology.

Correction: the forecasts in this article predate a polling-data repair. Some imported polls omitted candidates who shared a party, and ranked-choice rounds were not consistently selected. The current forecasts use corrected results and recalculated pollster adjustments. See current forecasts.

One new poll in California's 48th District shows a close race. SurveyUSA, polling for KGTV-TV and the San Diego Union-Tribune, surveyed 522 likely voters from August 27 to September 1 and found Democrat Marni von Wilpert at 45 percent and Republican Jim Desmond at 42 percent. Three points, in a sample that size, is not enough to separate them.

Our model has adjusted. It now gives von Wilpert a 26 in 100 chance of winning the seat, up from 26 percent's predecessor of 16 percent in the run completed at 8:32 a.m. ET on September 4. The mean margin moved from R+9.7 to R+6.3. The 5th-to-95th percentile range narrowed slightly on the Republican end and widened on the Democratic end, from R+25.8-to-D+6.2 to R+22.2-to-D+9.6. Desmond remains the favorite in our estimate, at 74 percent.

That is a real move, and it rests on a single survey. The forecast still sits meaningfully to the right of the poll itself, which is what a model built on more than one input does with one new data point: it takes some of it, not all of it. Our national generic ballot average, 25 polls over the past 45 days, stands at D+6.3, which is close to the district margin our model now expects here in the opposite direction.

For the House overall, our model gives Democrats a 93 percent chance of winning a majority, with a mean of 239.5 seats against the 218 needed for control. One caveat is ours to own: our simulation draws correlated error nationally, by census division, by state and by district, but not across demographically similar districts in different places, because we have no cluster data for it. That makes the House control number marginally firmer than a fuller treatment would leave it.

Polls this dispatch was written from

  • SurveyUSA · Sep 1, 2026 · n=522 · CA-48

Every constant behind these numbers is on the methodology page.